Upload
vuongcong
View
214
Download
0
Embed Size (px)
Citation preview
379
THE RISE – MATERIAL BEHAVIOUR IN GENERATIVE DESIGN
Martin Tamke, David Stasiuk, Mette Ramsgard ThomsenCentre for Information Technology and Architecture (CITA)
ABSTRACT
The research-based installation, The Rise, is led by the concept of a growing architecture able to
sense and dynamically adapt to its environment as it grows into form while continuously reacting
to its own material performance and behavioural constraints. This process is enabled through
the careful integration of digital simulation techniques with multi-hierarchical generative design
approaches. Aggregations of variably sized bundles of rattan core multiply, bend, branch and re-
combine into a distributed assembly that manifests an alternative to traditional structural systems.
The hybrid approach links a material system with simulation and the iterative generation of geom-
etry through a process of calibration at different stages of design. The project leverages emerging
computational strategies for growth in a model for an architectural practice that engages the com-
plexity and interdependencies that characterise a contemporary design practice.
1 The Rise – installed in the atrium space of the Fondation EDF Espace in Paris 2013
STRUCTURES ACADIA 2013 ADAPTIVE ARCHITECTURE 380
Phototropism in a plant
INTRODUCTION
The role of representation and prototyping is challenged by recent
developments wherein material performance operates as a key
driver of architectural design. Traditionally, the design process
focuses on dead geometry, as (Evans) stated, and any investiga-
tion of the interrelations and behaviour of environment, energy
and material—which may be considered through intuition during
the initial design process—are generally relegated to a secondary
post-generative analysis. In this process, a parameter space is
developed into which a generative model is located in pursuit of
an appropriate design solution expressed in a geometrical form.
The performance and behaviour of this model is then tested in a
physical manifestation or a digital simulation. Repeatedly cycling
through this process of parameterization, generation, simulation
and analysis enables the designer to optimize for selected effects
and characteristics of the model (Kolarevic). Often in fact, the
designer will deploy evolutionary algorithms to systematically in-
terrogate this parameter space with the goal of optimizing one or
more variables toward a pre-determined value. A potential short-
coming in this approach lies within the length of the feedback
loop created. Parametric design tools are practical but they effec-
tively rely on the definition of specific base infrastructures that
are resistant to change during the design process and over the
lifetime of a model (Scheurer). In order to establish a stable base
for the needed feedback loops it is essential to have models with
fixed topologies or rigid initial parameter states. Hence, attempts
to model for the changing state of both model and environment
and finding appropriate design solutions face the fundamental
limits of the computational models that underlie contemporary
practice. How can the rigidity and overly deterministic behaviour
of our models be overcome, and how can a more direct feedback
mechanism become an integral part of them?
Natural plant growth presents one diagrammatic framework for
developing a modelling approach that is able to adapt to internal
and external changes. Plant formation is accretive, and successive
growth iterations depend on the physical properties of previously
accumulated matter in relation to both itself and its environment.
The mechanisms that describe the motivators and geometric
principles of plant morphogenesis are known as tropisms, which
is derived from the Greek “Tropos”, meaning a “turning”. It was
first proposed in 1927 in the Cholodny–Went model, and is today
widely used to describe the continuous change and transfor-
mation expressed during growth. As described by Esmon et al
(2005), “plant tropisms are operationally defined as differential
growth responses that reorient plant organs in response to di-
rection of physical stimuli.” Examples of tropisms include those
reacting to light inputs (Figure 2), gravity, electrical, chemical or hy-
drological environments or touch. In plants, the biological mech-
anism for differentiation during formation is understood to be in
most instances auxin, a hormone that directs new cellular growth
and coordinates the emergence of the plant’s geometry. Within the
plant’s metabolism, the presence or absence of auxin triggers the
distribution of those available resources required for growth. The
Rise investigates how this diagram of growth exhibited by living
vegetative systems can inform computational models in the cre-
ation of a self-propagating structure. We ask whether strategies for
growth can operate as drivers of a generative system that is aware
of both its environment as a driver of algorithmic form generation,
and itself as a driver of material constraint and behavioural charac-
teristics. The installation is commissioned by the Fondation EDF
and is especially fabricated for the exhibition “Alive – Designing
with Living Systems” that takes place from April–September 2013
in Paris. The installation operates on an architectural scale and is
site specific to the gallery atrium at the Fondation EDF’s ”Espace”,
reaching from the ground to the ceiling.
The primary conceptual driver for the project is the idea of a
growing architecture: similar to a plant, the installation model
possesses a series of internally coded algorithmic responses that
guide branching and growth through material accumulation, orien-
tation and distribution.
As a designed system The Rise can combine behaviours of differ-
ent vegetative systems, one of these is self-grafting. Although this
behaviour is often associated with human intervention (Figure 3a),
it also occurs naturally in some plants, such as certain species of
ficus and fig (Figure 3b). In such instances, branches can meet and
grow together in new circular relationships that enable structural
performance. In this way, the installation is activated as a diagram
of natural growth processes and mimics nature’s manner of cre-
ating structural performance through strategies of redundancy
and material distribution, reinforcing these traits with the more
unusual hybrid growth methodology of self-grafting (Figure 4).
2
381
Natural self-grafting plant formations
Natural self-grafting plant formations
The installation model and assembly logics also emulate the fi-
brous material lamination mechanisms exhibited in tree growth. In
trees, growth occurs through the layering of multiple types of long
and fibrous cells on the perimeter of the trunk, which are arrayed to
achieve an integrated structural and nutrient distribution network.
During normal growth along the trunks and branches, these long
cells are generally arrayed parallel to growth direction with some
variation in favour of helical growth occurring in certain species at
both early and late-age growth. At branching moments, however,
these fibres are cross-laminated to provide multi-directional resis-
tance in the formation of a natural moment joint (Figure 5).
Rattan core is the primary material utilized in The Rise, as con-
struction material and as well as inspiration for the programming
of the behaviour of the artificial system. Rattan Core wood is a
vine-like palm that grows up host trees in constant search for light
in the dense jungles of Southeast Asia. The material used for the
installation is sustainably harvested in Malaysia, where the wood
is knife-extruded into round sections (the installation relies on
thicknesses of 5 mm, 10 mm and 19 mm). The woody material is
soft and is comprised almost entirely of a collection of continuous
and tightly packed hollow fibres. These can be differentiated from
other long woody plants like bamboo, which instead are regularly
broken into linear segments by a series of bracing nodes that
enable self-support. This organization strategy is tied to a rattan’s
reliance on other plants for structure, and it allows the rattan to
more efficiently transport water and nutrients from the ground
along winding branches for rapid growth and its ability to bend in
extreme angles towards the sun. This flexibility also affords the
material improved resistance to breaking, and its long fibres en-
dow it with high tensile performance. These characteristics—re-
silience to breaking, ease in bending, and tensile strength—are all
maintained post-harvest and make rattan a preferred material for
furniture and basket making in South Asia and popular throughout
the world. In its natural habitat rattan is structurally dependent on
the structures of other trees, and it grows through a process of
closely hugging its hosts.
Through a process of tight bundling along a normal trunk and
branch growth, and multi-directional active bending collections
at each branching moment, the rattan core is used in The Rise
in such a way as to mimic the behaviours of cellular vegetative
growth. Crucially, the highly flexible nature of rattan exhibits
bending behaviours that facilitate both fabrication as well as the
calibration of the digital design system in its continuous sim-
ulation of material characteristics during morphogenesis. The
model’s dynamic topology is then directed to branch and multiply
according to its internally encoded logics as a response both to
its environment, but furthermore in direct response to its own
behaviours in light of continuous accretion and simulated material
behaviour. Through these logics the installation explores the use
3b
3c
THE RISE – MATERIAL BEHAVIOUR IN GENERATIVE DESIGNTAMKE, STASIUK, THOMSEN
Self-grafting through human intervention3a
STRUCTURES ACADIA 2013 ADAPTIVE ARCHITECTURE 382
of highly redundant and distributed systems as an alternative to traditional structural systems. Instead
of leveraging the post-formation optimization of certain elements, we grow a multitude of intersecting
members that combine to form an integrated structural network. In this vision architecture is not a
static formalist proposition, but is instead continuously adapting to the dynamics of its surroundings
and its own material constraint while it grows into form.
SYNTHESIS OF ALGORITHM AND SIMULATION
The installation’s model emerges from an adaptive growth algorithm that combines the simulation of
plant growth logics with an active and continuously running particle-based physics simulation system.
These elements work in concert to activate the characteristics of the rattan core as primary contrib-
utors to form generation, and to enable an algorithmic procedure that is informed by both self and
environmental awareness. The Rise has been developed to operate as a climbing growth organism.
It utilizes structural “shoots” and gripping feet that identify and cling to their physical surroundings as
vine-like plants do and achieves structural triangulation through a process of self-grafting.
The Rise relies heavily on emulating a system of plant-like tropisms as a series of diagrammatic drivers
in form generation and performance. For our modelling process, we focused on only three types of
tropisms: phototropism (light-driven response), geotropism (gravity-driven response) and thigmotro-
pism (touch-driven response). In plant formation, these responses are registered through the triggering
of a hormone such as auxin. These hormones will often suppress (or activate) growth in certain cells
or locations by restricting (or promoting) energy-based resources in order to allow for an uneven and
ultimately directed cellular growth. Again via Esmon (2005): “Each tropic response has its own special
suite of molecules that are necessary for proper signal perception, signal amplification and attenuation
and elaboration of the growth response. While the establishment of hormone gradients is a required
step in each response, it’s not the hormone that does the dirty work. Auxin, for example, acts indi-
rectly through many different proteins to induce a growth response.” This localized cellular variation
serves to orient the plant according to an internalized rule system that ultimately produces goal-oriented
growth and organizational emergence. In our model we interpret tropisms through the algorithmic
Scaled model from the design process speculating algorithmic growth and networked behaviour through branching and self-connection
A simplified section of a branch reveals its complex fibrous setup, which works in compression as well as in tension
4 5
383
deployment of directional orientation and task assignment during
branching and those further topological transformations intro-
duced in the form of the self-grafting and climbing behaviours
described above.
In addition to this application of tropism-like characteristics, our
model also relies heavily on the diagrammatic simulation of light
providing energy for growth activation, both in terms of orientation
(phototropism) and substance (material accretion). To achieve this,
our model then uses a series of energy variables that are distrib-
uted according to a local branch’s proximity to a light source: the
closer the growth tip of a branch is to the light, the more energy
it is given for the three distinct growth processes of lengthening,
further branching and flowering. Energy is portioned out and in-
crementally accumulated for each of these actions by growth tip
according to parameter settings assigned by the designer, and at
each growth phase each tip tests the stores of energy is has avail-
able to execute any individual action. The threshold for accretion,
for example, is set lower than that for branching which in turn System for continuous growth filtered through the results of an on going simula-tion of physical properties during morphogenesis
7
Diagrammatic foundation of plant tropisms for algorithmic goal-oriented model coding6
THE RISE – MATERIAL BEHAVIOUR IN GENERATIVE DESIGNTAMKE, STASIUK, THOMSEN
STRUCTURES ACADIA 2013 ADAPTIVE ARCHITECTURE 384
is set lower than that for flowering. If a growth tip achieves the
energy required for branching, it passes any other accumulated
energy on to its child growth tips. Most crucially, the modelling of
tropisms and simulation of energy is tightly coupled with an on-
going physics-based simulation of geometric bending and twist-
ing under self-weight. This endows the growth model with the
highly pliable rattan material characteristics during its formation.
Since the model is allowed to embody the characteristics of an ac-
tive-bending structural system during its compilation, it emerges
as much as a non-deterministic material response to its own ge-
ometry as it does to the integral formational algorithms that com-
prise the rules for its modular accretion. Each successive phase of
growth is then actuated through the accumulation of new energy
and the model’s reading of both environment and its continuously
moving place in it, due to both material aggregation and dynamic
bending behaviour.
Careful management of model topology is instrumental for all
aspects of the recursive growth algorithm as well as for later use
in the production of detailing and fabrication systems, it is also
essential to the embedded continuous physical simulation. The
model exhibits its growth through the accretion of minimally trian-
gulated modules, which are managed in a mesh (Figure 8) and that
make up the core of the computational mechanism. Each open
face of the mesh is identified as a growth tip that has the capacity
to be activated by multiple drivers. The three points along each tip
define a local coordinate space that serves as the nexus for several
operations. Included among others are proximity testing mecha-
nisms for the simulation of energy acquisition and for the interpre-
tation of thigmotropism through grafting and climbing. The growth
tip coordinate systems also provide the directional sensing neces-
sary for multiple operations. New branches are assigned roles as
either light-seeking elements or structure seeking elements based
Minimally triangulated modular accretion and branching logics for management of spring-based system and growth/ branching topologies
Hierarchical assembly - standard units that combine to produce irregular structures based on hierarchical assembly
8 9a
9b
Bending under self-weight as a function of growth over time captured
on their orientation in their environment. As a growth tip branches
it forms a tetrahedron, the new open faces of which provide the
base point for each new branch and its local coordinate system
used for the orientation testing in this role assignment.
The mesh is managed such that its point, edge and face topology
is registered and deployed in a custom-written particle-based
spring and gravity simulation system. Early iterations of the model
sought to embed distinct bending and spring forces in this sys-
tem, but the constraints of this approach were revealed in its
failure to realistically account for torsion and orientation during
branching. The triangulated system developed in response to this
accommodates not only the orientation and modular accretion
tactics used for growth and tropism interpretation, but its truss-
like organization also results in a configuration that allows for a
wholly spring-based approach that simulates active bending, the
retention of orientation during branching and torsional behaviours.
385
Therefore, this mesh becomes a means for full model integration:
it combines the simulated physical performances with multiple
custom classes used for capturing vital information about the
model’s growing, branching, grafting and flowering events. The
spring-based system also endows the model geometry with direct
indicators of material performance. For this installation, measure-
ment of spring lengths versus their baseline as they react to new
algorithmic growth module accretion generates the data regarding
deformation in both tension and compression which is crucial to
the sizing and distribution of rattan members during detailing and
fabrication (Figure 10).
Each growth iteration can then be identified by a three-fold
cycle that includes: (a) environmental sensing in the form of
orientation according light-source location and gravity, as well
as proximity feeling for both itself and its physical surroundings,
(b) local measurement and distribution of accumulated light
energy for direct growth, branching and flowering, and (c) the
simulation of physical form and material performance as a result
of accretion. So rather than sequencing parameterization, gen-
eration, simulation and analysis into a series of discrete events,
The Rise seeks to collapse this process into a tightly integrated
synthetic whole.
DEVELOPMENT OF A BESPOKE FIBROUS MATERIAL SYSTEM
In an analogue to the tree climbing rattan, The Rise utilizes its
environment for structural support. In difference to rattan The
Rise can graft to itself. This creates redundant triangulation which
enables it to manifest a more dynamic geometry as it increases
overall structural strength. In its built instantiation, The Rise grows
from two seed points at the columns of the gallery and bends
under its own weight.
Overall generation with registration of absolute stresses derived during formation in the spring-based simulation engine, and local assignment of total material thicknesses according to this registration
10
The assembly logic is expressed as a series of struts and connec-
tion nodes. Each strut is comprised of a tightly packed bundle of
variably-sized rattan elements that behave as primary structural
members and tie related connection nodes together. Each con-
nection node is formed using the active bending properties of its
composite rattan members in opposition to one another. Variable
sizing in each member, according to their topological arrange-
ment, is a registration of the orientation derived from the digital
growth and simulation model (Figure 11).
As such the material’s systematized elastic deformation (active
bending) is an integral part of the system’s architecture (Lienhard,
2012), and is used actively both in the creation of branching points
and as a calculated part of the member sizing and distribution.
Embedded information about the material’s minimum allowable
radius and bending behaviours couples with orientation and defor-
mation information provided by the simulation to dynamically size
members at each node and along each strut.
Along the struts, the efficiency of the bundling behaves like a tex-
tile in that it relies on friction between fibres. In The Rise, the nec-
essary cohesion between fibres is created both through a strategy
for maximizing the continuity of individual rattan members (Figure
12) through multiple connection nodes and through a series of lo-
cally customized compression rings—or “packing nodes”—made
from CNC milled HDPE. Although a matrix material such as in
wood or composite fibre products (Nicholas 2012) would afford a
higher degree of structural efficiency, the taken approach consid-
ers the constraints of on-site installation and the engagement of
the fibres with the connection nodes.
Branching is used in vegetative systems to maximize coverage for
solar gain. For this biomimetic installation, it is then essential that
The Rise to grow towards the sun. Branching moments also define
weak points within the structure of a vegetative system. To combat
this, plants develop highly tensile and elastic joints between the
branch and the trunk. Where a crotch might seem to be a simple
split, woody plant growth patterns create interwoven fibres that pro-
vide stiffness and elasticity in multiple directions. This cross weav-
ing inspired the nodes of The Rise, where fibres are interwoven
through the distinct connections between each member incoming
to each branch node. Here variably sized and accumulated oppos-
ing rattan members in each node define the resulting orientation
of each outgoing strut. These opposing rattan members in each
connection node are organized again by a series of locally custom-
ized CNC milled HDPE elements—called stars—except that these
work not as compression rings but rather as organizing and guiding
systems for reorienting and managing the complex rattan member
topologies that emerge from the meeting of multiple struts.
These stars are designed along with the packing nodes as an inte-
grated system (Figure 13), where complex topological relationships
are registered and managed both within each connection node’s
THE RISE – MATERIAL BEHAVIOUR IN GENERATIVE DESIGNTAMKE, STASIUK, THOMSEN
STRUCTURES ACADIA 2013 ADAPTIVE ARCHITECTURE 386
set of stars and between each connection node through each
strut’s packing node system. Through these systems, information
derived from the growth/simulation system regarding geometry
and material thicknesses is registered through the installation’s
fabrication logic. Individual rattan members are allowed to bundle
and laminate in diagrammatic emulation of the same bundling and
lamination that occurs in woody plant fibres that grow in response
to their environment.
SYSTEM CALIBRATION
The development of the material system and its generative mod-
elling process ran in parallel. A series of material tests used to en-
gage in physical generative modelling techniques and to measure
the rattan’s bending capacities, and prototypes operated as inves-
tigations into the nature of the rattan’s behaviour. These tests yield
crucial data related to a number of parameters that informed the
model’s growth algorithms, physical simulations and final detailing
processes. Included here are the measurements of bending load
capacities, the composition of different sized members into the
tightly-bent formations that define the model’s branching node ge-
ometries, as well as analysis of the dynamic and time-dependent
features related to the minimal bending radii allowed in variable
material thicknesses.
In the case of the modelling process for The Rise, calibration of
both the simulation in terms of bending performance as well as the
fabrication system, member number and thickness is informed by
the anticipated load but as well by the desired angles within this
truly three-dimensional node. Observation of actual material config-
urations directly informed the system developed for calculating the
material composition of each opposing connection strand in any
given element. For example, for any four-point node topology, there
are a total of six topological connections (from points 0 to 1, 0 to 2,
0 to 3, 1 to 2, 1 to 3, and 2 to 3). Within each node, those two ele-
ments that have been grown with the least bending are afforded the
highest contribution of available material, such that their bending
strength is greater than their neighbours (Figure 11 and Figure 14).
In conjunction with this, the total material in each node is dictated
by the amount of deformation any node is required to resist, as
anticipated through the simulation-based modelling process.
Therefore, the total amount of material designated for any given
node is distributed across each topological connection possibility
according to both the total amount of resistance required of a node
and the local geometric demands of each topological connection’s
angle. All of these tactics were fully informed by the empirical ob-
servations of the rattan in the various material tests executed.
DISCUSSION AND CONCLUSION
The Rise successfully creates an integral and tightly looped re-
lationship between simulation and generation in architectural
design. The collapse of the space between generation, analysis
Synthesis of local active bending deformation with connection node branch orientations into member sizing and orientation for fabrication
Generated detailed model demonstrating dynamic continuity of fibres
Topologies between separate star elements through a packing node (star|pack-ing node|star) topology, with member continuities
11
12
13
387
and feedback distinction reduces the distinction between each
to a brief moment of internalized computation. With such a tight
feedback loop in place, a system awareness emerges that is able
to directly couple generative strategies with material behaviour
and consequence such that subsequent accretion and organiza-
tion directly relies on accumulated geometry and performance
characteristics. The use of this relatively simple spring-based
approach for the generation of the geometry of an installation is
sufficient for a proof-of-concept, but there is undoubtedly a great
deal of further research that can be pursued to locate a broader
base of simulation system inputs within iterative growth cycle
and continuous remapping of the parameter space. Although the
approach taken in the programming of the system shows a great
amount of resilience towards changes in the environment as to
the underlying assumptions and constructions on the material
level, further opportunities exist to more directly couple fabrica-
tion logics into the generative design process itself. It may be
possible that further considered use of the material, geometric
and organizational properties in such a hybrid system could op-
erate as further direct inputs in the refinement of the generative
modelling space.
The Rise (Figure 15) demonstrates as well that the integration of
feedback needs to take place on several levels and at different
time steps during design development. While the integration of
the described calibrated simulation of general bending behaviour
is effective in this design generation process, further integration
steps and research development will be necessary to determine
the viability of such an approach on a detailed design level.
Opportunity for further integration lies in a better translation
between material testing and the calibration of the simulation
system. System-wide bending behaviours and those related to op-
positional geometries along each connection are extracted from
these parallel physical modelling processes. Meanwhile, these are
wholly empirical in the sense that observation is translated into
both simulation model and final fabrication configuration based
on a series of localised tests. There is significant space to better
code local physical characteristics into the simulation system as
a means to further refine the relationship between digital feed-
back and anticipated physical performance. In its current state,
The Rise relies heavily upon approximation and global behaviour
parameterization. Therefore, while the local properties are de-
termined according to their differentiated responses and defor-
mations within this global setup, it should be possible to refine
the model such that these properties themselves are set during
simulation according to a more localized feedback system. It is
only through continued iterations in the development of such an
integrated simulation systems that an adequate level of precision
for design on an architectural scale can be established. This step
currently exceeds the capacity of the physics-based system as
currently configured and utilised.
Where the design with generative systems can only take place
through the configuration of the framing conditions and assump-
tions for the internal rule set (Tamke 2010), the use of simulation
in a design integration requires the use of adequate system cali-
bration. This system calibration was characterised in the case of
The Rise through a series of feedback loops between assumption
and experiment, with each new insight leading to new questions
and refined assumptions. Where the process of the calibration of
simulation is well understood, standard and absolutely necessary
in other disciplines (Winsberg 2010) it challenges the architectural
profession and the scale of its engagement. The Rise shows here
how the integration of the development of the digital model and
the physical systems allow for the cross seeding of information
and the mutual increase of knowledge.
14
15
Render highlight of variably-sized oppositional bending elements defining the geometry of a connection node
Detail of assembled structure with dark guideposts instrumental for the assembly of the structure
THE RISE – MATERIAL BEHAVIOUR IN GENERATIVE DESIGNTAMKE, STASIUK, THOMSEN
ACADIA 2013 ADAPTIVE ARCHITECTURESTRUCTURES 388
This knowledge might give the system more abilities or deter-
mine its behaviour in a better way. The integration of strategies
of growth mean on the other side that the system cannot be
informed about an optimal state it uses moreover its resilience to
adapt its growth to a changing environment which includes the
system itself. This cannot be understood as a neutral entity in an
environment. In contrary it is active and is as much influenced
by an examination of its own state as by changes of the environ-
ment, hence a preconceived optimization goal is not permitted.
The Rise can therefore only be understood as a speculation of a
more integrated future for design.
ACKNOWLEDGMENTSThe work would not have been possible without the enduring
efforts from Hollie Gibbons and Shirin Zhagi on design and devel-
opment, Hasti Valipor for the Parisian support, Carole Collet for
the invitation to the exhibition and excellent curatorship and Mette
Ramsgaard Thomsen for the guidance and support.
WORKS CITEDBall, Philip. 2009. Nature’s patterns: A tapestry in three parts. Oxford: Oxford University Press.
Esmon, C. A., Pedmale, U., and Liscum, E., “Plant tropisms: providing the power of movement to a sessile organism”, The International Journal of Developmental Biology 49: 665-674, 2005
Evans, Richard. 1995. The projective cast: architecture and its three geometries. Cambridge, Mass.: MIT Press.
Kolarevic, Branko. 2005. Architecture in the digital age design and manufacturing. New York: Taylor & Francis.
Lienhard, Julian, Alpermann, Holger, Gengnagel, Christoph, Knippers, Jan 2012. Active Bending: A Review on structures where bending is used as a self formation process. Seoul: Proceedings IASS- APCS Symposium 2012.
Muller, M., Stam, J., James, D., Thurey, N., Real Time Physics Class Notes from the Siggraph Conference, 2008
Nicholas, Paul, Tamke, Martin, Ramsgaard Thomsen, Mette,, Jungjohann, Hauke,. Markov, Ivan. (2012) Graded Territories: Towards the Design, Specification and Simulation of Materially Graded Bending Active Structures. San Francisco: Proceedings of the Proceedings of the 32nd Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA): Synthetic Digital Ecologies [.
Runions, A., Fuhrer, M., Lane, B., Federl, P., Rolland−Lagan, A. and Prusinkiewicz, P. Modeling and visualization of leaf venation patterns, ACM Transactions on Graphics 24(3), pp. 702−711, 2005
Scheurer, Fabian. (2013). Encoding Design. in Peters, Brady, Peters, Terri (ed.). Inside Smart Geometry. London: Wiley.
Shigo, Alex. 1994. Tree Anatomy. Snohomish: Shigo & Trees Associated.
Tamke, Martin, Riiber, Jacob, Jungjohann, Hauke. (2010) Generated Lamella. in Proceedings of the 30th Annual Conference of the Association for Computer Aided Design in Architecture (ACADIA). New York.
Winsberg, Eric (2010) Science in the Age of Computer Simulation. Chicago: The University of Chicago Press.
MARTIN TAMKE is an associate professor at the Centre for
Information Technology and Architecture (CITA) in Copenhagen.
He is pursuing a design led research on the interface and
implications of computational design and its materialization. He
joined the newly founded research centre in 2006 and shaped
its design based research practice. Projects on new design
and fabrication tools for wood and composite production led
to a series of digitally fabricated demonstrators that explore an
architectural practice based on performative approaches and
engaged with bespoke material behaviour. His research in the
emerging field of digital production in building industry led to
several academic and industry based projects with extensive
funding. Currently he is acting head of CITA and involved in the
7th framework project DURAARK and the Danish funded 4 year
Complex Modelling research project.
METTE RAMSGARD THOMSEN is an architect
working with digital technologies. Her research centres on the
relationship between crafts and technology framed through
“digital crafting” as a way of questioning how computation, code
and fabrication challenge architectural thinking and material
practices. Mette is a professor at the Royal Danish Academy of
Fine Arts, School of Architecture in Copenhagen, where she heads
the Centre for Information Technology and Architecture (CITA).
DAVID STASIUK is an architect and PhD fellow at the
Centre for Information Technology and Architecture in Copenhagen.
His research investigating development strategies for emergent
parameterisation is a component of the Centre’s larger, multi-
pronged Complex Modelling project. His own work is particularly
focused on investigating the possibility for developing emergent
parameter spaces through the integration of simulation systems
with topological transformation. His professional work has focused
on bespoke detailing for advanced architectural geometries,
computational design implementation, and the deployment of
digital fabrication and documentation techniques, some of which
was presented during the Acadia 2012 conference.